Recurring Break Down of VFDs.

Thread Starter

Richie. K

Joined Apr 27, 2022
12
For quite a while I've been having a challenge with the variable frequency drives I use to run vibrating motors on some equipment here. The challenge is the drives keep failing every time. I've used different brands of Inverters (Toshiba, ABB and Omron) so I believe the problem does not have to do with the brand of inverters but probably there's something I may be missing.
For the Toshiba and ABB inverters I lost all the IGBTs, but with the Omron (Part ID : 3G3MX2-A4015-E)which I'm currently using trips off with an error code E30 (Drive Error). This error does not permit me to reset even after I cycle power without the load connected.

The motors are new and their insulation resistances are good. Insulation resistance of the cables are also good.
I have 4 A Circuit breaker between the main 3-phase supply and the inverter and and output reactor (rated for this specific Omron vfd) between the vibrating motor due to the length of my cable (about 60 meters) to reduce the harmonics on the output lines. Each motor is connected to an Inverter.

Please I'm open to all Ideas. Kindly help.
 

MaxHeadRoom

Joined Jul 18, 2013
25,227
Have you tried running the motors without load for a test?
Have you tried a auto-tune on them? Or at least set the parameters correctly?
What kind of wiring between VFD & motor? VFD shielded or conduit with discrete conductors?
 

strantor

Joined Oct 3, 2010
5,968
I can't think of the right words to Google (cyclic load, oscillating load, vibratory load, none of these are right) to find the application note I wished to show you, but this is a known issue you are experiencing and there are ways to mitigate it. I believe the problem is when the load is unbalanced, the VFD is trying to maintain a constant speed through wildly varying torque requirements, so it is constantly either operating in an overtorque situation or a regeneration situation. The there is much cycling between high current and over voltage condition, which puts a lot of stress on the transistors.

You should try to balance the load as best you can. Then run in torque mode instead of speed mode, so this "hunting" can be avoided. If it is still a problem then install a braking resistor to handle the overvoltage. If it is still a problem then oversize the VFD by 50% or more.

I recommend you consult with your VFD reps for more guidance. The app note I am thinking of, was published either by Yaskawa or ABB. Someone at either of those organizations will understand your application and be able to provide better advice than I can.
 

Thread Starter

Richie. K

Joined Apr 27, 2022
12
Have you tried running the motors without load for a test?
Have you tried a auto-tune on them? Or at least set the parameters correctly?
What kind of wiring between VFD & motor? VFD shielded or conduit with discrete conductors?
The first question, Yes we ran the motors on no load for a brief period about 30 minutes. We did not experience any trip or error.
But when ran with load it works fine for about a month then it goes bad.

I've not tried auto tune on them. But I'm pretty sure I set the parameters correctly.

For the wiring we ran three 14 AGW size conductors through a conduit to the motors (FLA of the motors are 1.19 Amps).
 

Thread Starter

Richie. K

Joined Apr 27, 2022
12
I can't think of the right words to Google (cyclic load, oscillating load, vibratory load, none of these are right) to find the application note I wished to show you, but this is a known issue you are experiencing and there are ways to mitigate it. I believe the problem is when the load is unbalanced, the VFD is trying to maintain a constant speed through wildly varying torque requirements, so it is constantly either operating in an overtorque situation or a regeneration situation. The there is much cycling between high current and over voltage condition, which puts a lot of stress on the transistors.

You should try to balance the load as best you can. Then run in torque mode instead of speed mode, so this "hunting" can be avoided. If it is still a problem then install a braking resistor to handle the overvoltage. If it is still a problem then oversize the VFD by 50% or more.

I recommend you consult with your VFD reps for more guidance. The app note I am thinking of, was published either by Yaskawa or ABB. Someone at either of those organizations will understand your application and be able to provide better advice than I can.
Thanks you.
I can relate to what u are saying because at a point we used a higher rated vfd (“5.5 kW”both Omron and Toshiba) to run the motors separately (each motor 0.5 kW) and they operated quite fine for a long while.
Eventually we lost their IGBTs along the line. The problem still persisted.
I’d really appreciate it if u could help me get the application note u are talking about. I believe it can help.
 

strantor

Joined Oct 3, 2010
5,968
I’d really appreciate it if u could help me get the application note u are talking about. I believe it can help.
I'll try, but this is very significant:

at a point we used a higher rated vfd (“5.5 kW”both Omron and Toshiba) to run the motors separately (each motor 0.5 kW) and they operated quite fine for a long while.
Eventually we lost their IGBTs along the line. The problem still persisted.
If this is happening even with a VFD rated 11x higher than the motor, then I am almost certainly wrong in my initial suspicion. I highly doubt this is due to the reasons I described earlier. That 5.5kW VFD should be capable of turning the motor to slag long before the transistors fail. And the bus capacitors in a VFD that size should be capable of gobbling up any amount of energy that a 0.5kW motor can regenerate.

I suspect some outside assassin now. Some other, larger, device connected near it may be generating nasty voltage spikes; this usually affect input diodes though, not so much output transistors. In any case, it wouldn't hurt to install an input reactor.

Does the failure always happen at the same time of day (ex: always on night shift)? Sometimes mains voltages sag/spike in conjunction with fluctuations in usage from nearby heavy consumers. If you're down the road from an electric arc furnace or something, and they shut it down, 460V nominal mains might briefly jump to 600V or so, before the power company's gear can make necessary corrections.

This is a vibrating bed; what is it vibrating? Some metallic/conductive dust? Are the VFDs installed inside a dustproof enclosure? Conductive dust will be drawn to the high current areas and short things out.

Are the VFDs exposed to high temperature? High humidity? Other ambient extremes?

Your situation must be explained by some external cause I think, not the application.
 

Thread Starter

Richie. K

Joined Apr 27, 2022
12
I'll try, but this is very significant:


If this is happening even with a VFD rated 11x higher than the motor, then I am almost certainly wrong in my initial suspicion. I highly doubt this is due to the reasons I described earlier. That 5.5kW VFD should be capable of turning the motor to slag long before the transistors fail. And the bus capacitors in a VFD that size should be capable of gobbling up any amount of energy that a 0.5kW motor can regenerate.

I suspect some outside assassin now. Some other, larger, device connected near it may be generating nasty voltage spikes; this usually affect input diodes though, not so much output transistors. In any case, it wouldn't hurt to install an input reactor.

Does the failure always happen at the same time of day (ex: always on night shift)? Sometimes mains voltages sag/spike in conjunction with fluctuations in usage from nearby heavy consumers. If you're down the road from an electric arc furnace or something, and they shut it down, 460V nominal mains might briefly jump to 600V or so, before the power company's gear can make necessary corrections.

This is a vibrating bed; what is it vibrating? Some metallic/conductive dust? Are the VFDs installed inside a dustproof enclosure? Conductive dust will be drawn to the high current areas and short things out.

Are the VFDs exposed to high temperature? High humidity? Other ambient extremes?

Your situation must be explained by some external cause I think, not the application.
The first question, Yes we ran the motors on no load for a brief period about 30 minutes. We did not experience any trip or error.
But when ran with load it works fine for about a month then it goes bad.

I've not tried auto tune on them. But I'm pretty sure I set the parameters correctly.

For the wiring we ran three 14 AGW size conductors through a conduit to the motors (FLA of the motors are 1.19 Amps).
Have you tried running the motors without load for a test?
Have you tried a auto-tune on them? Or at least set the parameters correctly?
What kind of wiring between VFD & motor? VFD shielded or conduit with discrete conductors?
I have configured the auto tune now, I'll monitor the operation of the vfd.
 

Thread Starter

Richie. K

Joined Apr 27, 2022
12
Thanks you.
I can relate to what u are saying because at a point we used a higher rated vfd (“5.5 kW”both Omron and Toshiba) to run the motors separately (each motor 0.5 kW) and they operated quite fine for a long while.
Eventually we lost their IGBTs along the line. The problem still persisted.
I’d really appreciate it if u could help me get the application note u are talking about. I believe it can help.
I have configured the vfd to run on torque mode. I'll keep monitoring to see the performance.
 

Thread Starter

Richie. K

Joined Apr 27, 2022
12
I'll try, but this is very significant:


If this is happening even with a VFD rated 11x higher than the motor, then I am almost certainly wrong in my initial suspicion. I highly doubt this is due to the reasons I described earlier. That 5.5kW VFD should be capable of turning the motor to slag long before the transistors fail. And the bus capacitors in a VFD that size should be capable of gobbling up any amount of energy that a 0.5kW motor can regenerate.

I suspect some outside assassin now. Some other, larger, device connected near it may be generating nasty voltage spikes; this usually affect input diodes though, not so much output transistors. In any case, it wouldn't hurt to install an input reactor.

Does the failure always happen at the same time of day (ex: always on night shift)? Sometimes mains voltages sag/spike in conjunction with fluctuations in usage from nearby heavy consumers. If you're down the road from an electric arc furnace or something, and they shut it down, 460V nominal mains might briefly jump to 600V or so, before the power company's gear can make necessary corrections.

This is a vibrating bed; what is it vibrating? Some metallic/conductive dust? Are the VFDs installed inside a dustproof enclosure? Conductive dust will be drawn to the high current areas and short things out.

Are the VFDs exposed to high temperature? High humidity? Other ambient extremes?

Your situation must be explained by some external cause I think, not the application.
If I may ask how will the input reactor help the input diodes please.

The failure does not happen at a specific time of the day. I've had situations where the failure occurred during both day and night shifts.
The motors are vibrating metallic screens (mesh) to sieve out particular sizes of carbon material for production. Yes environment where the motors are located is very dusty (carbon dust which is highly conductive). But the VFDs are located in an enclosed cabinet in the motor control center (MCC). Chances of dust getting to the cabinet are minimal.
The VFDs aren't exposed to high temperatures or humidity. But the entire area outside has a lot of carbon dust around.
 

Thread Starter

Richie. K

Joined Apr 27, 2022
12
It knocks down spikes in the incoming voltage that might be caused by nearby high power devices. spikes that could damage the VFD.
Ok I’ll try that as well. Is there anything else u’ll suggest I can consider, maybe any other protection and why?
 

shortbus

Joined Sep 30, 2009
9,375
The motors are vibrating metallic screens (mesh) to sieve out particular sizes of carbon material for production. Yes environment where the motors are located is very dusty (carbon dust which is highly conductive).
Is it a VFD problem or is the vibrating motor the problem? You haven't said what type of motor, and I like others are thinking these are the sealed vibrators that are called vibrating motors. Is this what you have or are these regular motors with a crank driving the screens?
 

Thread Starter

Richie. K

Joined Apr 27, 2022
12
Is it a VFD problem or is the vibrating motor the problem? You haven't said what type of motor, and I like others are thinking these are the sealed vibrators that are called vibrating motors. Is this what you have or are these regular motors with a crank driving the screens?
I'm facing the challenge on the VFD. The motors are in good condition actually.
Yes its a sealed vibrator.

Invicta vibrator motor
Model: BL 25
FLA: 1.19A
2 pole.
400 V
 

LowQCab

Joined Nov 6, 2012
2,060
"" and and output reactor (rated for this specific Omron vfd) between the vibrating motor due to the length of my cable (about 60 meters) to reduce the harmonics on the output lines. ""

What does this "Reactor" consist of exactly ?
And why do You think that You need it ?
.
.
.
 

MaxHeadRoom

Joined Jul 18, 2013
25,227
I always figure that the 3phase rectifier and Very large bank of capacitors on the input offers quite a degree of protection over spikes.
According to one course I took on VFD's the input filter is more often seen useful for PF correction than suppression.
 

strantor

Joined Oct 3, 2010
5,968
I always figure that the 3phase rectifier and Very large bank of capacitors on the input offers quite a degree of protection over spikes.
According to one course I took on VFD's the input filter is more often seen useful for PF correction than suppression.
Right, that is how I figure it too. VFD manufacturers often recommend them if more than one VFD is installed. Servo manufacturers usually recommend them for all installations. I rarely ever install them unless there is a problem or good reason to do so, and I think this qualifies.
 

GetDeviceInfo

Joined Jun 7, 2009
2,096
I had an old vfd blow fuses at inconvenient times, until I installed an input choke that alleviated that problem.

Are you braking these motors?
 
Top